[v2,1/3] KVM: arm64: Add helper vgic_write_guest_lock()
Commit Message
Currently, the unknown no-running-vcpu sites are reported when a
dirty page is tracked by mark_page_dirty_in_slot(). Until now, the
only known no-running-vcpu site is saving vgic/its tables through
KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_SAVE_TABLES} command on KVM device
"kvm-arm-vgic-its". Unfortunately, there are more unknown sites to
be handled and no-running-vcpu context will be allowed in these
sites: (1) KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_RESTORE_TABLES} command
on KVM device "kvm-arm-vgic-its" to restore vgic/its tables. The
vgic3 LPI pending status could be restored. (2) Save vgic3 pending
table through KVM_DEV_ARM_{VGIC_GRP_CTRL, VGIC_SAVE_PENDING_TABLES}
command on KVM device "kvm-arm-vgic-v3".
In order to handle those unknown cases, we need a unified helper
vgic_write_guest_lock(). struct vgic_dist::save_its_tables_in_progress
is also renamed to struct vgic_dist::save_tables_in_progress. Besides,
"asm/kvm_mmu.h" needs to be included for "vgic.h" for the definition
of kvm_write_guest_lock().
No functional change intended.
Suggested-by: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Gavin Shan <gshan@redhat.com>
---
arch/arm64/kvm/vgic-sys-reg-v3.c | 1 +
arch/arm64/kvm/vgic/vgic-irqfd.c | 1 +
arch/arm64/kvm/vgic/vgic-its.c | 13 +++++--------
arch/arm64/kvm/vgic/vgic-mmio-v2.c | 1 +
arch/arm64/kvm/vgic/vgic-mmio.c | 1 +
arch/arm64/kvm/vgic/vgic-v4.c | 1 +
arch/arm64/kvm/vgic/vgic.c | 1 +
arch/arm64/kvm/vgic/vgic.h | 13 +++++++++++++
include/kvm/arm_vgic.h | 2 +-
9 files changed, 25 insertions(+), 9 deletions(-)
Comments
Hi Gavin,
On Fri, Jan 20, 2023 at 07:44:03AM +0800, Gavin Shan wrote:
> Currently, the unknown no-running-vcpu sites are reported when a
> dirty page is tracked by mark_page_dirty_in_slot(). Until now, the
> only known no-running-vcpu site is saving vgic/its tables through
> KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_SAVE_TABLES} command on KVM device
> "kvm-arm-vgic-its". Unfortunately, there are more unknown sites to
> be handled and no-running-vcpu context will be allowed in these
> sites: (1) KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_RESTORE_TABLES} command
> on KVM device "kvm-arm-vgic-its" to restore vgic/its tables. The
> vgic3 LPI pending status could be restored. (2) Save vgic3 pending
> table through KVM_DEV_ARM_{VGIC_GRP_CTRL, VGIC_SAVE_PENDING_TABLES}
> command on KVM device "kvm-arm-vgic-v3".
>
> In order to handle those unknown cases, we need a unified helper
> vgic_write_guest_lock(). struct vgic_dist::save_its_tables_in_progress
> is also renamed to struct vgic_dist::save_tables_in_progress. Besides,
> "asm/kvm_mmu.h" needs to be included for "vgic.h" for the definition
> of kvm_write_guest_lock().
>
> No functional change intended.
>
> Suggested-by: Oliver Upton <oliver.upton@linux.dev>
> Signed-off-by: Gavin Shan <gshan@redhat.com>
> ---
> arch/arm64/kvm/vgic-sys-reg-v3.c | 1 +
> arch/arm64/kvm/vgic/vgic-irqfd.c | 1 +
> arch/arm64/kvm/vgic/vgic-its.c | 13 +++++--------
> arch/arm64/kvm/vgic/vgic-mmio-v2.c | 1 +
> arch/arm64/kvm/vgic/vgic-mmio.c | 1 +
> arch/arm64/kvm/vgic/vgic-v4.c | 1 +
> arch/arm64/kvm/vgic/vgic.c | 1 +
> arch/arm64/kvm/vgic/vgic.h | 13 +++++++++++++
> include/kvm/arm_vgic.h | 2 +-
> 9 files changed, 25 insertions(+), 9 deletions(-)
You wouldn't have to add the include all around the shop if you instead
just stuck it in vgic.h...
Having said that, we really ought to get a fix in for this sooner rather
than later. I just hit it myself testing kvmarm/next.
Marc, could you take care of the include fix when applying?
Hi Oliver,
On 1/27/23 7:20 AM, Oliver Upton wrote:
> On Fri, Jan 20, 2023 at 07:44:03AM +0800, Gavin Shan wrote:
>> Currently, the unknown no-running-vcpu sites are reported when a
>> dirty page is tracked by mark_page_dirty_in_slot(). Until now, the
>> only known no-running-vcpu site is saving vgic/its tables through
>> KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_SAVE_TABLES} command on KVM device
>> "kvm-arm-vgic-its". Unfortunately, there are more unknown sites to
>> be handled and no-running-vcpu context will be allowed in these
>> sites: (1) KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_RESTORE_TABLES} command
>> on KVM device "kvm-arm-vgic-its" to restore vgic/its tables. The
>> vgic3 LPI pending status could be restored. (2) Save vgic3 pending
>> table through KVM_DEV_ARM_{VGIC_GRP_CTRL, VGIC_SAVE_PENDING_TABLES}
>> command on KVM device "kvm-arm-vgic-v3".
>>
>> In order to handle those unknown cases, we need a unified helper
>> vgic_write_guest_lock(). struct vgic_dist::save_its_tables_in_progress
>> is also renamed to struct vgic_dist::save_tables_in_progress. Besides,
>> "asm/kvm_mmu.h" needs to be included for "vgic.h" for the definition
>> of kvm_write_guest_lock().
>>
>> No functional change intended.
>>
>> Suggested-by: Oliver Upton <oliver.upton@linux.dev>
>> Signed-off-by: Gavin Shan <gshan@redhat.com>
>> ---
>> arch/arm64/kvm/vgic-sys-reg-v3.c | 1 +
>> arch/arm64/kvm/vgic/vgic-irqfd.c | 1 +
>> arch/arm64/kvm/vgic/vgic-its.c | 13 +++++--------
>> arch/arm64/kvm/vgic/vgic-mmio-v2.c | 1 +
>> arch/arm64/kvm/vgic/vgic-mmio.c | 1 +
>> arch/arm64/kvm/vgic/vgic-v4.c | 1 +
>> arch/arm64/kvm/vgic/vgic.c | 1 +
>> arch/arm64/kvm/vgic/vgic.h | 13 +++++++++++++
>> include/kvm/arm_vgic.h | 2 +-
>> 9 files changed, 25 insertions(+), 9 deletions(-)
>
> You wouldn't have to add the include all around the shop if you instead
> just stuck it in vgic.h...
>
> Having said that, we really ought to get a fix in for this sooner rather
> than later. I just hit it myself testing kvmarm/next.
>
> Marc, could you take care of the include fix when applying?
>
I've posted v3 to have a separate PATCH[1/4] where the header file inclusions
are handled, to save Marc's valuable time. After 'kvm_mmu.h' is included to
'vgic.h', the duplicate inclusions of 'kvm_mmu.h' needs to be removed. A separate
patch would make the follow-up patches clean.
https://lore.kernel.org/kvmarm/20230126235451.469087-1-gshan@redhat.com/T/#t
Thanks,
Gavin
@@ -7,6 +7,7 @@
#include <linux/kvm.h>
#include <linux/kvm_host.h>
#include <asm/kvm_emulate.h>
+#include <asm/kvm_mmu.h>
#include "vgic/vgic.h"
#include "sys_regs.h"
@@ -7,6 +7,7 @@
#include <linux/kvm_host.h>
#include <trace/events/kvm.h>
#include <kvm/arm_vgic.h>
+#include <asm/kvm_mmu.h>
#include "vgic.h"
/**
@@ -2187,7 +2187,7 @@ static int vgic_its_save_ite(struct vgic_its *its, struct its_device *dev,
((u64)ite->irq->intid << KVM_ITS_ITE_PINTID_SHIFT) |
ite->collection->collection_id;
val = cpu_to_le64(val);
- return kvm_write_guest_lock(kvm, gpa, &val, ite_esz);
+ return vgic_write_guest_lock(kvm, gpa, &val, ite_esz);
}
/**
@@ -2339,7 +2339,7 @@ static int vgic_its_save_dte(struct vgic_its *its, struct its_device *dev,
(itt_addr_field << KVM_ITS_DTE_ITTADDR_SHIFT) |
(dev->num_eventid_bits - 1));
val = cpu_to_le64(val);
- return kvm_write_guest_lock(kvm, ptr, &val, dte_esz);
+ return vgic_write_guest_lock(kvm, ptr, &val, dte_esz);
}
/**
@@ -2526,7 +2526,7 @@ static int vgic_its_save_cte(struct vgic_its *its,
((u64)collection->target_addr << KVM_ITS_CTE_RDBASE_SHIFT) |
collection->collection_id);
val = cpu_to_le64(val);
- return kvm_write_guest_lock(its->dev->kvm, gpa, &val, esz);
+ return vgic_write_guest_lock(its->dev->kvm, gpa, &val, esz);
}
/*
@@ -2607,7 +2607,7 @@ static int vgic_its_save_collection_table(struct vgic_its *its)
*/
val = 0;
BUG_ON(cte_esz > sizeof(val));
- ret = kvm_write_guest_lock(its->dev->kvm, gpa, &val, cte_esz);
+ ret = vgic_write_guest_lock(its->dev->kvm, gpa, &val, cte_esz);
return ret;
}
@@ -2743,7 +2743,6 @@ static int vgic_its_has_attr(struct kvm_device *dev,
static int vgic_its_ctrl(struct kvm *kvm, struct vgic_its *its, u64 attr)
{
const struct vgic_its_abi *abi = vgic_its_get_abi(its);
- struct vgic_dist *dist = &kvm->arch.vgic;
int ret = 0;
if (attr == KVM_DEV_ARM_VGIC_CTRL_INIT) /* Nothing to do */
@@ -2763,9 +2762,7 @@ static int vgic_its_ctrl(struct kvm *kvm, struct vgic_its *its, u64 attr)
vgic_its_reset(kvm, its);
break;
case KVM_DEV_ARM_ITS_SAVE_TABLES:
- dist->save_its_tables_in_progress = true;
ret = abi->save_tables(its);
- dist->save_its_tables_in_progress = false;
break;
case KVM_DEV_ARM_ITS_RESTORE_TABLES:
ret = abi->restore_tables(its);
@@ -2792,7 +2789,7 @@ bool kvm_arch_allow_write_without_running_vcpu(struct kvm *kvm)
{
struct vgic_dist *dist = &kvm->arch.vgic;
- return dist->save_its_tables_in_progress;
+ return dist->save_tables_in_progress;
}
static int vgic_its_set_attr(struct kvm_device *dev,
@@ -10,6 +10,7 @@
#include <kvm/iodev.h>
#include <kvm/arm_vgic.h>
+#include <asm/kvm_mmu.h>
#include "vgic.h"
#include "vgic-mmio.h"
@@ -12,6 +12,7 @@
#include <kvm/iodev.h>
#include <kvm/arm_arch_timer.h>
#include <kvm/arm_vgic.h>
+#include <asm/kvm_mmu.h>
#include "vgic.h"
#include "vgic-mmio.h"
@@ -9,6 +9,7 @@
#include <linux/irqdomain.h>
#include <linux/kvm_host.h>
#include <linux/irqchip/arm-gic-v3.h>
+#include <asm/kvm_mmu.h>
#include "vgic.h"
@@ -11,6 +11,7 @@
#include <linux/nospec.h>
#include <asm/kvm_hyp.h>
+#include <asm/kvm_mmu.h>
#include "vgic.h"
@@ -131,6 +131,19 @@ static inline bool vgic_irq_is_multi_sgi(struct vgic_irq *irq)
return vgic_irq_get_lr_count(irq) > 1;
}
+static inline int vgic_write_guest_lock(struct kvm *kvm, gpa_t gpa,
+ const void *data, unsigned long len)
+{
+ struct vgic_dist *dist = &kvm->arch.vgic;
+ int ret;
+
+ dist->save_tables_in_progress = true;
+ ret = kvm_write_guest_lock(kvm, gpa, data, len);
+ dist->save_tables_in_progress = false;
+
+ return ret;
+}
+
/*
* This struct provides an intermediate representation of the fields contained
* in the GICH_VMCR and ICH_VMCR registers, such that code exporting the GIC
@@ -263,7 +263,7 @@ struct vgic_dist {
struct vgic_io_device dist_iodev;
bool has_its;
- bool save_its_tables_in_progress;
+ bool save_tables_in_progress;
/*
* Contains the attributes and gpa of the LPI configuration table.